Roles of motor proteins in neurons

National Heart, Lung, And Blood Institute

Investigators

Linked publications & trials

Abstract

Myosin 10 is required for normal cerebellar development, cerebellar function, and Purkinje neuron structure and function. The Purkinje neuron (PN) is the master neuron of the cerebellum, as it receives all inputs into the cerebellar cortex, is the sole output from the cortex, and is essential for coordination, balance and learning precise motor tasks. We showed previously that myosin Va transports tubules of ER into PN spines to promote synaptic plasticity and motor learning (Wagner et al, Nat. Cell Biol. 2011), and that myosin 18A targets the guanine nucleotide exchange factor -Pix to PN spines to promote spine maturation (Alexander et al, FASEB J 2021). Here we describe ongoing efforts using to define the function of myosin 10 (Myo10) in PNs, which are unique among CNS neurons in possessing very high levels of this filopodial myosin, and in undergoing filopodia-to-spine conversion without prior innervation. At the whole organ level, six week-old Myo10 knockout (KO) mice (Heimsath et al, Sci. Rep. 2017) exhibit cerebellar hypoplasia and misshapen and/or missing cerebellar lobes. Additionally, Calbindin staining of cerebellar slices from these mice reveals defects in the alignment of PN soma and in the orientation of PN dendritic arbors within the molecule layer. Importantly, Calbindin staining of cerebellar slices from mature, six month-old Myo10 KO mice reveals major defects in PN morphology, including reduced dendritic arborization and greatly reduced spine density. Consistent with these observations, Myo10 KO mice exhibit significant defects in cerebellar function (e.g. maintaining balance). Finally, GFP-tagged Myo10 expressed in cultured PNs localizes dramatically at the tips of filopodia-like extensions at the leading edge of forming neurites, and the miRNA-mediated knockdown of Myo10 in these cells results in defects in spine maturation and cell polarity (reduced dendritic arborization, increased number of axons). Together, these results argue that Myo10 is required for normal cerebellar development, cerebellar function, and PN structure and function, and they pave the way for future efforts designed to identify the molecular mechanisms by which this MyTH4/FERM myosin promotes these processes. Myosin Va is not required for myelination. Dilute lethal mice, which are homozygous for the myosin Va functional null allele dl20j, exhibit a profound neurological phenotype characterized by tonic-clonic seizures, opisthotonos and ataxia. These symptoms emerge about P12 and result invariably in death at about P21. Previous work from our lab showed that myosin Va transports tubules of endoplasmic reticulum into the dendritic spines of cerebellar Purkinje neurons to promote synaptic plasticity and motor learning (Wagner et al, Nat. Cell Biol. 2011). While this function can explain, at least in part, the ataxia exhibited by dilute lethal mice, it cannot explain the breadth and severity of their symptoms. Relevant to this, seizures, muscle spasms and ataxia are exhibited by several mouse myelination mutants, as well as by people with certain demyelinating diseases. Consistently, Noguchi et al. (Exp. Neur., 1983) reported that the activity of 2, 3-cyclic nucleotide 3phosphodiesterase (CNPase), a myelin-associated enzyme expressed exclusively by myelin-producing oligodendrocytes, is greatly reduced in the brains of P20 dilute lethal mice. Moreover, Sloane and Vartanian (J. Neuroscience, 2007) reported that the maturation of oligodendrocyte precursors into mature oligodendrocytes in vitro is inhibited by expressing a myosin Va dominant negative construct, as well as by microinjecting a myosin Va antibody. These authors also provided EM evidence that the corpus callosum, optic nerve and spinal cord of P15 dl20j/dl20j mice contain fewer myelinated axons than the same tissues from P15 WT mice. Finally, an interaction screen performed using our myosin Va TAP-tagged mouse (Alexander et al, Cytoskeleton, 2019) showed that myelin basic protein (MBP) and CNPase are enriched in the final, tandem affinity tag eluate. Together, these results are consistent with idea that myosin Va plays a significant role in myelination by promoting the differentiation of oligodendrocyte precursor cells (OPCs) into mature oligodendrocytes (OLs), and/or by supporting the synthesis of myelin sheaths by mature OLs. That said, the identification of defects of any sort in dilute lethal mice close to their time of death raises questions regarding cause-effect given that these mice are very sick (dehydrated, emaciated, hypoxic, etc.). Given this, we sought here to obtain more definitive information regarding a role for myosin Va in myelination. Antibody staining of WT OPCs as they differentiate in vitro into mature OLs shows that myosin Va levels remain essentially constant over this 5-day process, suggesting that the myosin functions in both developing OPCs and mature OLs. More importantly, dishes in which equal numbers of WT or dl20j/dl20j OPCs were plated showed the same number of myelin basic protein (MBP)-positive, mature OLs at day 5, arguing that myosin Va is not required for the differentiation of OPCs into OLs. Similarly, the densities of PDGFR-positive OPCs and CC1-positive OLs in sagittal sections of corpus collosum from day-12 WT and dilute lethal mice are the same. With regard to the ability to make myelin, WT and dl20j/dl20j OLs seeded on nanowires to mimic axons produced the same length myelin sheaths on these wires. Moreover, the signals for MBP in sagittal sections of day-12 cerebelli from WT and dl20j/dl20j mice were indistinguishable. Finally, MBP-stained electron micrographs of corpus collosum and optic nerves from day-12 and day-18 dl20j/dl20j mice showed no reduction in myelination at day-12 and only a small reduction at day-18 relative to WT tissues. Together, our results show that myosin Va is not required for the differentiation of OPCs into mature OLs or for the ability of mature OLs to create myelin sheaths. Our results also argue that the profound neurological phenotype exhibited by dilute lethal mice is most likely due to defects in the function of one or more neuronal cell types. Efforts to unravel this will be aided by our recent creation of a conditional myosin Va knockout mouse.